Research Units: Bocconi University (Dirk Hovy), George Washington University (Rebekah Tromble), TU Delft, Liverpool University, Syracuse University
The project proposes to help Twitter tackle two significant problems affecting the health of conversations on its platform: echo chambers and online abuse. Given the concerns about growing polarization and the spread of misinformation we will develop four metrics to quantify healthy conversations and measure their effectiveness when applied as machine learning tools. The first two metrics mutual recognition and diversity of perspectives will help Twitter diagnose issues that arise when users isolate themselves from those who hold differing opinions. The second two metrics incivility and intolerance will help Twitter identify and address abuse and targeted harassment. In order to classify these measures at scale we draw upon existing work in a variety of computational fields notably natural language processing and network analysis but take this work further in addressing the metrics outlined here. Moreover beyond merely detecting and measuring mutual recognition diversity of perspectives incivility and intolerance we propose to study the effects these four phenomena have on users. In doing so we offer a theoretically and empirically driven approach that will help Twitter diagnose the relative health of the conversation on its platform.